MiR-450 inhibits cell proliferation, migration and invasion and induces apoptosis in gastric cancer via targeting CREB1

Background (cid:0) Gastric cancer seriously affects human health and research on gastric cancer is increasing. In recent years, molecular targets are hot research topics. We aimed to explore the effects and mechanisms of miR-450 on the development and progression of gastric cancer. Methods (cid:0) We used gain-of-function approaches to investigate the gastric cancer cell proliferation, apoptosis, migration and invasion, including, RT-qPCR, CCK-8, colony formation, ow cytometry, western blot, wound healing, transwell chamber, HE, TUNEL, dual luciferase reporter and tumor formation analysis. Results (cid:0) We found that the expression levels of miR-450 were greatly decreased in gastric cancer cells and overexpression of miR-450 inhibited the gastric cancer cell proliferation, migration and invasion, while induced apoptosis in gastric cancer in vitro. Moreover, we demonstrated that ectopic expression of miR-450 inhibited tumor growth in vivo. At the molecular level, overexpression of miR-450 signicantly increased the expression levels of apoptosis-associated proteins, including Caspase-3, Caspase-9 and BAX, while inhibited the expression level of Bcl-2. Mechanically, luciferase reporter experiment suggested that CREB1 had a negative correlation with miR-450 expression and knockdown of CREB1 alleviated gastric cancer. Furthermore, we also found that miR-450 inhibited the activation of AKT/GSK-3β signaling pathway to inhibit the progression of gastric cancer. Conclusions induced AKT/GSK-3β which may provide a new


Background
Human gastric cancer is one of the common malignancies around the world [1]. Approximately 850,000 cases of gastric cancer are diagnosed and 650,000 deaths occur each year [2]. Different treatment strategies, including surgery, chemotherapy, and radiation, have achieved remarkable advance. However, overall therapeutic activity for advanced disease remains poor [3]. Thus, it is critical to acquire a better understanding of molecular mechanisms and new treatment strategies urgently need to be developed for improving the treatment of gastric cancer patients.
MicroRNAs (miRNAs) are a kind of endogenous small RNA with a length of about 20-24 nucleotides, which have a variety of important regulatory roles in cells. New research showed that miRNAs were involved in the development and regulation of gastric cancer. MiR-4317 inhibits cell proliferation and blocked the conversion of S-G2/M in gastric cancer cells, promising a therapeutic molecular target [4]. MiR-455 suppresses cell proliferation of gastric cancer and migration by modulating EGFR, acting as a potential target for treatment of gastric cancer [5]. MiR-582-5p inhibits cell proliferation and promoted apoptosis by regulating AKT3 [6]. Other studies have also shown the inhibitory effect of miRNAs on the progression and development of gastric cancer, including miR-744, miR-140-5p, miR-181a, miR-182, miR-802, miR-21, miR-149 [7][8][9][10][11][12][13]. On the other hand, miR-450, as a novel miRNA, which function as a tumor suppressor in the regulation of glaucoma [14]. However, its speci c role in the gastric cancer and the underline mechanism remain unknown.
Based on its anti-cancer role previously, we hypothesized that miR-450 may act as a tumor-suppressor to regulate the development and progression of gastric cancer. In the present study, we found that the expression levels of miR-450 were signi cantly decreased in gastric cancer cells. Overexpression of miR-450 inhibited cell proliferation, migration, and invasion and induced cell apoptosis in gastric cancer in gastric cancer in vitro. Moreover, overexpression of miR-450 inhibited the growth of tumor in vivo. Furthermore, we found that CREB1 served as the molecular target and mediated all the anti-tumor effects of miR-450. Taken together, these results indicates that miR-450 may is a potential molecular target in the treatment of gastric cancer.

Cell culture
The human gastric cancer cell lines (BGC-823, SGC7901) and human gastric epithelial cell (GES-1) were purchased from the Cell Bank of the Chinese Academy of Science (Shanghai, China). Cells were maintained in Dulbecco's modi ed Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS, Gibco, NY, USA) in a humidi ed air at 5% CO 2 atmosphere at 37℃. RNA extraction and real-time quantitative PCR (RT-qPCR) RNA was isolated by Trizol (Invitrogen, CA, USA) according to the manufacturer's protocols. RNAs were reverse transcribed to cDNA by employing PrimeScript RT Master Mix (Takara, Dalian, China) following the manufacturer's protocol. PCR ampli cation was conducted with the SYBR Premix Ex TaqTM Kit (Takara, Dalian, China). GAPDH and U6 were used to normalize the expression. The relative expression was calculated using the 2 −ΔΔCt method.

Oligonucleotides and transfection
The plasmid carrying the CREB1 CDS domain was used to overexpress the CREB1 in gastric cancer cells and a pcDNA (pcDNA3.1) vector was used as a negative control (GenePharma, Guangzhou, China). The cDNA encoding CREB1 CDS domain was ampli ed by PCR, and then subcloned into the pcDNA3.1 vector (Invitrogen, CA, USA) to obtain the pcDNA-CREB1. The miR-450 mimic and the negative control (NC) were synthesized by GenePharma (Guangzhou, China). The transfection was performed by using Lipofectamine 3000 Reagent (Life Technologies, Carlsbad, CA, USA).

Cell proliferation assay
Cell Counting Kit-8 (CCK-8) and colony formation assays were performed to examine gastric cancer cell proliferation. In brief, BGC-823 or SGC7901 cells (1 × 10 4 /well) were seeded in 96-well plates and seeded for 0 h, 24 h, 48 h and 72 h, respectively. Then BGC-823 or SGC7901 cells were incubated with 10 µL CCK-8 solution for 4 h at 37℃. The optical density (OD) was recorded using a microplate at 490 nm (Multiskan MK3, Thermo Scienti c, USA). For colony formation assay, BGC-823 and SGC7901 cells (2 × 10 4 cells/well) were seeded in 24-well plates. After incubation for 12 days, the cells were immobilized with paraformaldehyde for 30 min, and stained with 10% crystal violet for another 30 min. Colonies were counted and photographed with a light microscope (Olympus, Tokyo, Japan).

Cell apoptosis assay
Flow cytometry was conducted to investigate the gastric cancer cell apoptosis. Cells (1 × 10 4 cells/well) were seeded in six-well plates and culture for 48 h. Thereafter, cells were washed using PBS, treated with 5 µL of annexin V-FITC and 5 µL of PI in the dark for 15 min at room temperature. Then apoptosis cells were detected through a FACScalibur Flow Cytometry (BD Biosciences, CA, USA).

Cell migration and invasion assays
Wound healing and transwell chamber (5-µm pore size, Costar, Cambridge, MA, USA) assays were conducted to examine gastric cancer cell migration and invasion abilities. For wound healing assay, cells were seeded into six-well plates. The supernatant uid was removed when BGC-823 or SGC7901 cells were highly con uent (> 90%). Scratches were made using a sterile pipette tip, with the scratch width remaining the same. After continuous culture for 48 h, the width of the scratch was photographed and recorded under a microscope (⋅ 100). For invasion assay, BGC-823 or SGC7901 cells were estimated by transwell assays. In short, transfected BGC-823 or SGC7901 cells were added to the upper chamber loaded with matrigel (Corning, Cambridge, MA) and the bottom of the chamber were supplemented with complete medium containing 1% FBS. 48 h later, cells on the surface of membranes were wiped out. Invaded cells were xed in 10% formaldehyde, dyed with 0.1% crystal violet and then counted with a light microscope (Olympus, Tokyo, Japan).

Hematoxylin-eosin staining (HE) assay
Tumor slices were stained with hematoxylin for 5 min, then rinsed for 1 min, and returned to blue by 1% ammonia (30 s). Afterwards, slices were ushed with running water (1 min). Furthermore, slices were stained by 0.5% H&E (for 1 min), rinsed (for 30 s), made into transparent, and nally mounted with neutral gum.

TUNEL assays
The slices were washed by PBS and then immobilized for 30 min with 4% paraformaldehyde. After washed with PBS once, the slices were added with 0.1% Triton X-100 for 2 min, and then washed with PBS once. Afterwards, 3% H 2 O 2 was used for incubation (5 min). Then the slides was rinsed, and maintained with 50 µL TUNEL at 37 °C overnight. After rinsed by PBS again, the TUNEL reaction was visualized by chromogenic staining with DAB (Sigma-Aldrich). Light microscope was applied to observe the slides and the TUNEL-positive BGC-823 or SGC7901 cells were calculated in Image J software.
Dual luciferase reporter assay A putative 3'-untranslated regions (3'-UTR) of CREB1 was mutated using mutagenesis kit (Promega, USA). Wild type and mutant sequences were ampli ed and inserted into the vector to construct luciferase reporter plasmids according to the manufacturer's instructions (Promega, USA). The luciferase activities were detected with the dual luciferase reporter kit (Promega, USA). Luciferase activity was measured by dual-luciferase reporter assay system (Promega) and presented as re y luciferase intensity normalized to Renilla luciferase activity.

Western blotting analysis
Total proteins in tissues or cells were dissolved with RIPA lysate buffer (Beyotime Inc, Shanghai, China). The protein was quanti ed using BCA protein assay Kit (Thermo Scienti c, CA, USA). The protein samples were separated by 12% polyacrylamide gel, which were transferred to the PVDF membrane, sealed with 5% skim milk powder. The membrane was incubated with the primary antibody at 4℃overnight and then incubated with HRP coupled secondary antibody (Santa Cruz Inc, CA, USA) at room temperature for 1 h. The protein signal was detected by ECL detection reagents (Thermo Scienti c, CA, USA) and GAPDH worked as the internal reference.

Xenograft tumors in nude mice
Female nude mice (6-week-old, 18-22 g) were provided by Nanjing Medical University and housed under germ free conditions. Animal care and use were carried out according to the ethical guidelines by the First A liated Hospital of USTC Animal Care and Use Committee. Nude mice were maintained in a 12 h light/12 dark cycle in a temperature-and humidity-controlled environment. To detect the effect of miR-450 on tumor growth in vivo, BGC-823 cells (1 × 10 6 cells) were injected subcutaneously into the right axilla of the nude mice. Following a 30-day period, nude mice were sacri ced, and neoplasms were isolated for further analyses. Note that the tumor volumes were recorded every week and calculated with the formula: Volume = 0.5 × length × wide 2 .

Statistical analyses
All data are presented as the mean ± SD. Each bar expressed the mean ± SD of three independent experiments. Statistical signi cance between two or multiple groups was analyzed by t-test or one-way ANOVA using SPSS 17.0 (SPSS Inc, Chicago, IL, USA). Experiments were repeated three times independently. Statistical signi cance was assumed when P < 0.05.

Results
Effect of miR-450 on gastric cancer cell proliferation and apoptosis RT-qPCR was employed to screen the expression level of miR-450 in gastric cancer cells. As shown in Fig. 1A, the expression levels of miR-450 were signi cantly decreased in gastric cancer cells (BGC-823 and SGC7901), compared with that in human gastric epithelial cell (GES-1). To further evaluate the roles of miR-450 in the regulation of gastric cancer, gain-of-function assays were implemented in BGC-823 and SGC7901 gastric cancer cells, respectively. The e ciency of transfection was validated by RT-qPCR (Fig. 1B). More importantly, the results of CCK-8 and colony formation assays indicated that the cell viability and number of colonies were signi cantly decreased in BGC-823 and SGC7901 cells which transfected with miR-450 mimic, compared with that in NC mimic group (Fig. 1C, 1D). Subsequently, ow cytometry assay were conducted to examine cell apoptosis of gastric cancer. We found that cell apoptosis rate was signi cantly increased in miR-450 mimic group, compared with NC mimic group (Fig. 1E). Consistent with these ndings, western blot analysis revealed that the expression levels of apoptosis-associated proteins, including Caspase-3, Caspase-9 and Bax, were signi cantly increased, whereas Bcl-2 was decreased in miR-450 mimic-transfected BGC-823 and SGC7901 cells (Fig. 1F). These results indicated that miR-450 overexpression inhibited cell proliferation and induced cell apoptosis of gastric cancer.

Overexpression of miR-450 inhibits tumor growth in vivo
To con rm whether miR-450 inhibit the tumor growth of gastric cancer in vivo, we generated BGC-823/miR-450 mimic cells and their negative control, then injected them into nude mice. Representative images of tumor in the nude mice with miR-450 mimic and NC mimic ( Figure. 2A). Tumor volume and tumor weight were signi cantly decreased in miR-450 mimic group compared with that in NC mimic group ( Figure. 2B, 2C). Furthermore, HE and IHC assays showed that the cell proliferation gene of Ki67 was signi cantly inhibited in miR-450 mimic group compared with that in NC mimic group. Furthermore, the results of TUNEL assay indicated that the cell apoptosis was signi cantly promoted in miR-450 mimic group compared with that in NC mimic group (Figure. 2D). Consistent with these ndings, western blot assay suggested that the expression levels of apoptosis-associated proteins, including Caspaes-3, Caspase-9 and Bax, were signi cantly decreased in miR-450 mimic group compared with that in NC mimic group, contrary by Bcl-2 ( Figure. 2E). These results indicated that overexpression of miR-450 suppressed tumor growth in vivo.

overexpression of miR-450 inhibits gastric cancer cell migration and invasion
Cell migration and invasion play important role in the pathogenesis of cancer metastasis. The results of wound healing assay showed that the wound closure rate was signi cantly decreased in miR-450-overexpressed BGC-823 and SGC7901 gastric cancer cells, respectively, compared with that in NC mimic group (Fig. 3A). Moreover, the results of transwell invasion assay showed that ectopic expression of miR-450 limited the inventory capability of BGC-823 and SGC7901 cells, compared with that in NC mimic group (Fig. 3B). Furthermore, western blot assay suggested that the expression levels of migration-and invasion-associated proteins, miR-450 mimic obviously decreased the levels of MMP-2 and MMP-9 in both BGC-823 and SGC7901 cells, compared with that in NC mimic group (Fig. 3C). These data suggested that overexpression of miR-450 inhibits migration and invasion of BGC-823 and SGC7901 cells.

MiR-450 targets and negatively regulates CREB1
The expression levels of CREB1 were signi cantly increased in BGC-823 and SGC7901 gastric cancer cells (Fig. 4A). To study on the possible targets of miR-450 involved in the occurrence of gastric cancer, rstly, TargetScan was carried out to predict that CREB1 was potential candidate of miR-450. We found that CREB1 might be a candidate target of miR-450 (Fig. 4B). Dual luciferase reporter assay was conducted in HEK-293T cells. The luciferase reporter assay showed that miR-450 mimic repressed the relative luciferase activities containing the WT 3'-UTR of CREB1, but had no obvious effect on Mut 3'-UTR of CREB1 (Fig. 4C). Moreover, RT-qPCR and western blot analysis were adapted to evaluate the expression of CREB1 in BGC-823 and SGC7901 cells transfected with miR-450 mimic or NC mimic, and the data revealed that up-regulation of miR-450 decreased the expression of CREB1 (Fig. 4D, E). These results indicated that miR-450 targets and negatively regulates the CREB1.

Knockdown of CREB1 inhibits gastric cancer cell proliferation, migration, invasion and induces apoptosis,
After identifying CREB1 as a target of miR-450 in gastric cancer, we then focused on whether CREB1 regulate the progression gastric cancer. We evaluated the e ciency of knockdown of CREB1 in BGC-823 and SGC7901 cells by RT-qPCR (Fig. 5A). CCK-8 and colony formation assays showed that the cell viability and the number of colonies were signi cantly decreased in BGC-823 and SGC7901 cells in sh-CREB1 group compared with that in sh-NC group (Fig. 5B, 5C). Subsequently, ow cytometry was conducted to examine cell apoptosis after knockdown of CREB1 expression in BGC-823 and SGC7901 cells and the results indicated cell apoptosis rate was signi cantly increased in CREB1-knockdowned BGC-823 and SGC7901 gastric cancer cells, compared with that in NC mimic group (Fig. 5D). Wound healing and transwell chamber assays showed that the gastric cancer cell migration and invasion capability was signi cantly inhibited in CREB1-knockdowned BGC-823 and SGC7901 gastric cancer cells, compared with that in NC mimic group (Fig. 5E, 5F). These results showed that knockdown of CREB1 suppressed gastric cancer cell proliferation, migration and invasion and induced apoptosis.

CREB1 overexpression partially restores the effects of miR-450 on gastric cancer
To dissect the role of CREB1 in mediating the inhibitory effects of miR-450 on gastric cancer, overexpression CREB1 in gastric cancer cells transfected with miR-450 mimic. CCK-8 and colony formation assays indicated that the cell viability and the colony numbers of gastric cancer cells were signi cantly decreased in BGC-823 and SGC7901 gastric cancer cells in the miR-450 mimic-treated group, when compared with that in NC mimic + pcDNA 3.1 group, while CREB partially abolished the inhibitory effects of miR-450 mimic on cell proliferation of BGC-823 and SGC7901 gastric cancer cells (Fig. 6A and  6B). Flow cytometry assays indicated that miR-450 mimic-induced promotion of cell apoptosis was prominently abrogated by CREB1 overexpression (Fig. 6C).
At the molecular level, western blot assay indicated that the expression levels of apoptosis-associated proteins, including Caspaes-3, Caspase-9 and BAX, were signi cantly increased in BGC-823 and SGC7901 gastric cancer cells treated with miR-450 mimic, compared with that in NC mimic + pcDNA 3.1 group, whereas the overexpression of CREB similarly abolished the activation effects of miR-450 mimic on the expression of apoptosis-associated proteins, including Caspaes-3, Caspase-9 and Bax. Coincidence with these results, the expression levels of apoptosis-associated protein of BCL-2 were controversially changed (Fig. 6D). These results pointed out that CREB1 overexpression partially restored the effects of miR-450 on proliferation and apoptosis of gastric cancer.

MiR-450 regulates the gastric cancer progression through inhibiting AKT/GSK-3β signaling pathway
Given the importance of AKT/GSK-3β in the regulation of gastric cancer, thus we determined the protein expression levels of the AKT/GSK-3β pathway by western blot assay (Fig. 7). We found that the expression levels of the AKT/GSK-3β pathway-associated protein, including GSK-3β and AKT, had no signi cantly differences in all the experiments. However, the protein phosphorylation levels were markedly decreased in response to the overexpression of miR-450, while CREB overexpression correspondingly increased AKT/GSK-3β phosphorylation in miR-450 mimic-treated BGC-823 cells. In addition, to further identify the role of AKT/GSK-3β in mediating the anti-gastric cancer role of miR-450, we use a SC79 as an activator of the AKT/GSK-3β pathway. As shown in Fig. 8A, 8B, 8D and 8E, CCK-8 and colony formation, wound healing and transwell chamber assays showed that the gastric cancer cell proliferation, migration and invasion abilities were signi cantly decreased in the miR-450 mimic-treated group, while signi cantly increased in the miR-450 mimic + SC79 group compared with that in miR-450 mimic group. These results were further con rmed by ow cytometry, which showed that the apoptosis rate markedly increased in miR-450 mimic group, while signi cantly decreased in miR-450 mimic + SC79 group (Fig. 8C). The data suggested that miR-450 represses the gastric cancer progression by targeting CREB1 through inhibiting AKT/GSK-3β signaling pathway.

Discussion
Our ndings found that miR-450 was signi cantly decreased in gastric cancer. Functionally, overexpression of miR-450 inhibited cell proliferation, migration and invasion in gastric cancer in vitro, while facilitated cell apoptosis of gastric cancer in vitro. Over-expression of miR-450 suppressed the tumor growth in vivo. At the molecular level, overexpression of miR-450 increased the expression levels of apoptosis-associated proteins, including Caspase-3, Caspase-9 and Bax, while inhibited the expression level of Bcl-2. Moreover, overexpression of miR-450 suppressed the migration and invasion-associated protein expression levels, including, MMP-2 and MMP-9. Mechanically, we found that miR-450 alleviated the development and progression of gastric cancer via targeting CREB1 and regulated the AKT/GSK-3β signaling pathway. In conclusion, miR-450 was a new molecular target for the treatment of gastric cancer.
In recent years, researches pay great attention to gastric cancer, trying to nd an effective new method to treat gastric cancer. Feng et al. reported that miR-518 suppresses the progression of gastric cancer by promoting apoptosis via targeting MDM2 and may be a promising molecular target for treating gastric cancer for the time to come [15]. Liu et al. demonstrated that miR-204 modulates the EMT to inhibit the gastric cancer cell migration and invasion via regulating snai1in vitro and in vivo [16]. Wang et al. proves that miR-129-5p inhibited gastric cancer cell proliferation and EMT by HMGB1, acting as a potential target for curing gastric cancer [17]. Wang et al. testi es that overexpression of miR-128b inhibits cell proliferation, migration and invasion, and promoted apoptosis in gastric cancer cells via down-regulation of A2bR [18]. Wu et al. reports that up-regulation of miR-449c suppressed gastric cancer cell growth and promoted apoptosis, while down-regulation of miR-449c promoted gastric cancer cell growth and inhibited apoptosis [19]. Wang et al. demonstrates that over-expression of miR-217 inhibits gastric cancer cell proliferation, invasion and promotes apoptosis via regulating GPC5, which acts as a potential therapeutic target [20]. Besides, other miRNAs, for instance, miR-376c-3p [21], miR-133b [22], miR-99b-5p and miR-203a-3p [23], miR-491-5p [24], miR-143 [25] are intensively involved in the regulation of the development and progression of gastric cancer. Thus miRNAs is closely associated with the occurrence and development of gastric cancer. Here are some researches indicated that Comparing to the above studies, which suggested that miR-450 was even more speci c in gastric cancer.
MiR-21, miR-450 and miR-149 participating in the regulation of glaucoma have been reported [14]. Moreover, miR-21 and miR-149 are associated with the development of gastric cancer [12,13]. More importantly, given the glaucoma exhibited similar cancer type with gastric cancer, hence, we explored the roles of miR-450 in gastric cancer and miR-450 was con rmed to be a tumor suppressor in gastric cancer. As miR-21 and miR-149 play a role in other various cancers, for instance, colorectal cancer [26], ovarian cancer [27], breast cancer [28], bladder cancer [29], hepatocellular cancer [30], it is worthy to investigate whether miR-450 also plays a role in these cancers. In the following studies, we will further study the role of miR-450 in different cancers, so as to illustrate the role of miR-450 in various cancers, which exhibit the advantages of miR-450 as a molecular therapeutic target.
Researches have reported the AKT/GSK-3β signaling pathway is closely associated to the development and progression of gastric cancer. Hua et al. reported that knockdown of ANXA11 suppressed gastric cancer cell proliferation, invasion and migration through the AKT/GSK-3β pathway, which was acted as a prognostic factor and promising molecular target for gastric cancer treatment [31]. Pan et al. demonstrated that CD36 regulated palpitate acid-induced metastasis in gastric cancer by AKT/GSK-3β/βcatenin pathway, acting as a potential molecular target for clinical intervention [32]. In our study, we found that miR-450 blocked the AKT/GSK-3β signaling pathway to regulate the progression of gastric cancer and miR-450 levels were related with strength of the AKT/GSK-3β signal pathway.
In conclusion, miR-450 targets CREB1 and inactivates the AKT/GSK-3β signaling pathway to inhibit the development and progression of gastric cancer. miR-450 may act as a tumor-suppressor factor of gastric cancer, promising to be a new molecular target for the treatment of gastric cancer.  The experimental content meets ethical requirements and obtains permission.

Consent for publication
The authors declare that they have no con ict of interests.

Availability of data and material
The source of all data and information is true and reliable

Competing interests
The authors declare that they have no con ict of interests.

Funding
This work was supported by the Youth Fund Project of The First Hospital of USTC (West District, No. 2018YJQN002).

Con icts of interest
There are no con icts of interest.